Flipping the Photoswitch: Ion Channels Under Light Control 107
control of ion channel function, further design and application of PCLs was not
revisited until nearly four decades later. The first molecule to follow was 4-GluAzo,
a PCL designed for the Kainate receptor-subfamily (KAR) of tetrameric iGluRs.
With its lipophilic tail, 4-GluAzo resembles LY339434 and discriminates between
GluK1 and GluK2 subtypes in a heterologous expression system (Volgraf et al.
2007 ). 4-GluAzo is more active in its trans-isomer but exhibits markedly reduced
potency and efficacy compared to glutamate (Glu). A recent crystal structure of
the GluK2 ligand-binding domain (LBD) in complex with 4-GluAzo confirmed
several hypothesis made during the design of 4-GluAzo (Reiter et al. 2013 ). This
structure confirmed that the Glu moiety of 4-GluAzo indeed forms those contacts to
the ligand-binding pocket that were previously observed for Glu. The structure also
revealed that the lipophilic tail protrudes to the protein surface with the formation
of additional PCL-protein contacts. Finally, the structure allowed attributing the
reduced efficacy to a more limited conformational change of the GluK2 LBD that is
characteristic for partial agonists.
The clamshell-like LBD of iGluRs is of prokaryotic origin and shared amongst
the three major iGluR subfamilies (AMPA receptors (AMPAR), KARs and NMDA
receptors (NMDARs)) (Mayer 2011 ; Janovjak et al. 2011 ). However, sufficient
structural differences exist between the subfamilies and these differences prevented
the direct application or even modification of 4-GluAzo as a PCL of AMPARs and
NMDARs. To functionally discriminate between AMPAR and other iGluRs, Bn-
TetAMPA, a highly specific AMPAR agonist with a lipophilic tail was derivatized
into AB-tetrazolyl-AMPA-3 (ATA-3) (Stawski et al. 2012 ). Like 4-GluAzo, ATA-3
is functionally active in its dark-adapted trans-isomer. Both 4-GluAzo and ATA-3
are capable of triggering trains of action potentials (APs) in primary neurons by
activating endogenous iGluRs.
UV light, which is required for trans-cis-isomerization of unmodified AB, may
harm biological samples and exhibits limited depth of tissue penetration. An im-
portant advance in the design of PCLs is thus the development of molecules that
react to visible light, which can be achieved by increasing the electron density at
the AB using substituents. For instance, trans-cis-isomerization of ATA-3 occurs
readily in blue light because of an electron-donating para-dimethylamino substitu-
ent (Stawski et al. 2012 ). Notably, unlike unmodified AB that thermally relaxes to
its trans-isomer on the time scale of many minutes to hours, such modified ABs
relax within seconds. Similar improvements in wavelength sensitivity and thermal
relaxation were also achieved in PCLs of voltage gated ion channels and exploited
for restoration of retinal function (see Sect. 4.2).
A recent variation of PCLs encompasses molecules with the ability to tune the
conductance of ligand-gated ion channels rather than to directly control channel
opening or closing. Two potentiating PCLs have been developed for pentameric
GABAARs and applied in heterologous expression systems, primary cells and a
tadpole model system (Stein et al. 2012 ; Yue et al. 2012 ). AP2 and MPC088 are
both AB derivatives of propofol, a common amnestic agent and powerful posi-
tive allosteric GABAAR modulator. However, in contrast to MPC0888, in which
AB is conventionally coupled to propofol, a new approach was chosen in AP2 by